基于SVM-RFE的动力下肢假肢运动模式识别  

Motion Pattern Recognition of Power Lower-limb Prostheses Based on SVM-RFE

作　　者：韩松均 王志明[1] 寇旭 魏纪宗 Han Songjun

机构地区：[1]上海大学机电工程与自动化学院,上海200072

出　　处：《工业控制计算机》2023年第8期77-79,共3页Industrial Control Computer

摘　　要：下肢运动模式识别是动力型假肢控制系统最重要的一部分,为了充分利用截肢者大腿残肢运动信息和地形坡度信息,通过安装在残肢接受腔和足面的IMU以及足底压力传感器构成识别信号源,从中提取时域特征值用于运动模式识别。通过构建层次SVM识别模型,并采用RFE算法对子分类器选择合适特征,最终完成对于平地行走、上楼梯、下楼梯、上斜坡、下斜坡五种步态模式的识别。实验结果表明,该方法的识别准确率达到了99.5%,因此基于SVM-RFE的动力下肢假肢运动模式识别方法是有效可行的。Lower limb motion pattern recognition is the most important part of the control system of power prosthetics.In order to make full use of the movement information of amputees'thigh stump and topographic slope information,in this study,IMUs and plantar pressure sensors installed in the stump and foot are used as recognition signal sources,from which time-domain eigenvalues are extracted for motion pattern recognition.By constructing hierarchical SVM recognition model and using RFE algorithm to select suitable features for sub-classifiers,five gait modes of flat walking,up stairs,down stairs,up slope and down slope are finally recognized.Experimental results show that the recognition accuracy of this method is up to 99.5%.Therefore,the motion pattern recognition method based on SVM-RFE is effective and feasible.

关 键 词：动力下肢假肢 步态模式识别 支持向量机 

分 类 号：R318[医药卫生—生物医学工程] TN911.7[医药卫生—基础医学]